Split testing emails: 15 email elements to test

At the end of the second Lord of the Rings movie, our heroes take on the might of Sauron’s army in an epic battle of good vs evil. 

From the moment the fighting starts, the odds are stacked against them. We feel every blow from the Urak-hai. Mines, ladders and battering rams hammer the castle walls. An endless sea of Orcs crash against our frantic heroes.

Eventually, the exhausted, dwindling band of survivors begin to realise their desperation. They have run out of steam, no longer able to protect the innocents hiding in the castle. The inevitable, hopeless loss closes in.

And then, just as all is lost, the sun rises and we look to the East. Gandalf arrives to ride down into battle, turn the tide, and defeat the evil hoards.

Similarly epic battles are played out across the land between subject lines and CTAs. But instead of Middle Earth, the arena is email marketing testing.

Granted, these have fewer Dwarves being tossed. But I’ll bet if the Dark Lord knew he could get a higher click-through rate on his messages, perhaps he too would have pitched different email elements against each other instead of wasting his time trying to secure Helm’s Deep.

As a modern marketer, rather than the ruler of Modor, you know that the success of your email marketing campaigns can be improved by testing various elements. So let’s quickly cover off the what and the why of testing, before examining which email elements to test.

What is split testing in email marketing?

A/B testing, or split testing, is a way of improving your marketing by testing changes based on real-world results. For email marketing, split testing aims to find out what drives higher open rates and click-through rates.

In any A/B test, two variants are designed and delivered to half the audience. When A/B testing email elements, your email software will handle delivering the variants to the different cohorts.

There are several ways you can carry out the test. One option is to test your theory with a small chunk of your total list. Then you send out the winning email to your remaining subscribers.

The second option is to send each variant to 50% of the total audience in any week. You can then take the winning design into the following week, gradually building up your own best practice.

The last option is to test by changing elements each week. This method isn’t great practice for A/B testing. As multiple elements are changing, you can’t be sure which one caused any increase or decrease. However, if you lack the software or email list size, this can still deliver results over time.

Why should you split test emails?

OK, so it’s probably pretty obvious why you should test emails: to learn what works and to improve results. But if you want to demonstrate to your boss why you should be testing emails, here are a few email marketing statistics to help you.

A massive 87% of marketers use email marketing.

And 59% of them say email is the channel that sees the highest ROI.

It’s not surprising, as ROI for email marketing can be as high as £42 for every £1 spent.

But you need to get it right, which is where testing comes in.

And despite that, a massive 39% of companies don’t test or segment their emails.

15 email elements to test

So what can you test? Here are 15 ideas to get you started.

Subject lines

How does someone decide to open your email? The choice rests almost entirely on your email subject line. So it’s the perfect place to start testing.

Here are some tests to consider:

• Subject line length
• Is the subject a question?
• Starting the subject line with Fwd: or Re: 
• Capitalisation: all caps, title case, or all lowercase?
• Add some intrigue

We’ve previously written about the lessons we learned when we analysed four million subject lines from our prospecting emails.

Pre-header

The other way you can improve open rates is with your pre-header, so it’s also ripe for split testing.

An email pre-header is the line of text that is visible from the inbox. Think of it as the Goose to your email’s Maverick: sure it could do the job on its own, but it’s better to have a wingman.

From name

Sure you’ll be sending your emails from the same email address, but you can still get creative with who they are from.

As you can see in the screenshot above, emails can come from the company, a specific person, or a mix of both.

Content/theme

You can analyse the subjects or themes of your emails. Although many factors affect your metrics, if you have enough data you may spot patterns in which topics are successful.

Certain themes might emerge that resonate with your audience, and other topics that consistently deliver lower click through.

Blog post title

Your content team might have spent ages lovingly crafting the perfect headline for SEO, but will that title resonate with an email audience?

As long as it’s not misrepresenting the content of the post, amending a blog post title in the email can up the click-through rates.

Timing

Which day of the week should you send your emails for the best response. What’s the best time of day? Does it actually matter?

Yes, it does. Email campaign results can change massively depending on when they are sent. We know, we’ve analysed millions of emails to reveal when to send. 

Frequency

How often you send emails can be tested, and you might want to keep an eye on your unsubscribe rates as you increase the frequency of emails sent.

Greeting

Why not test your greetings? Hi, howdy, what’s up, hola, welcome… find out what gets people engaged with your email from the off. 

Just never say hello. It’s one of our golden rules.

Links vs. buttons

Do you paste links in to keep your email design functional and transparent? Or do you ask the design team to spend an afternoon creating a fancy button? And then test the colour, size and shape of that button over the following weeks?

You may think the button wins every time. You may also be wondering how it took your design team half a day to design a button. The key to A/B testing your emails is to question everything in order to learn something. The fancy-looking solution doesn’t always win.

HTML email vs plain text email

While we’re on the subject, are your emails sent out in a lovely crafted HTML design? Or do you keep things simple with a plain text email?

Again, you might naturally be drawn to the fancy looking solution, but it’s not necessarily the most effective. Testing HTML email against plain text can see some surprising results.

Number of links

How many links do you add to your emails? It can depend on a lot; if it’s a monthly round-up then it’s likely to include loads.

But in our tips on cold email copywriting, we’ve recommended dropping them all. What’s the answer? It depends. Which means, in keeping with this article, the answer is to test it.

Attachments

Our guide to writing prospecting emails claims you should never include attachments. We stand by that: when it comes to prospecting you want the reader to take one action alone – get in touch with you. 

What about other forms of email marketing? Often, you’ll want the person to come to your site to take an action, like downloading a whitepaper, so you can monitor success. That doesn’t mean you shouldn’t test. Remember, assume nothing.

Personalisation

Personalisation is a no brainer when it comes to emails. Nobody wants to feel like they are one of a million faceless consumers, endlessly trying to earn 15 million merits.

But you can test personalisation a lot. Even something as simple as a recipient’s name: do you add it to the subject line, the greeting, or somewhere in the email body? 

You can also personalise the content of the email, by segmenting your email list into different cohorts.

Call to action

The conversion rate will be influenced by every factor of your email, but the call to action is a huge one. Test, test, and test again.

Tone of voice

How do you approach email copy? Do you use email templates? Go for a human approach or corporate? Conversational or polished? Your tone of voice is one more email element that can be tested. 

Testing email elements can continually improve your results. If you’re looking for your emails to deliver leads right now, you might want to get in touch with us. We’re prospecting experts, with five years and millions of emails of testing data behind us.